Researchers Develop AI Detecting Alzheimer's via Language
Researchers at Stevens Institute of Technology developed an AI tool that diagnoses Alzheimer’s disease with 95% accuracy by analyzing patients’ spoken or written language patterns, the team reports. The convolutional neural network converts sentences into 512-dimensional vectors and combines these with expert-crafted features to identify pronoun substitution, awkward phrasing, and other markers, while producing human-readable explanations for clinicians and enabling extension to other languages and conditions.
Scoring Rationale
Solid university research with clear methods, limited by single-team reporting and lack of peer-reviewed clinical validation.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
